Myosin is a superfamily of prototypical molecular motors that play important roles in diverse biological processes ranging from vesicle trafficking, cell motility to muscle contractions and signal transductions. Although the functional cycle of myosins is understood in an out-line form, many detailed questions remain concerning the coupling between conformational properties of the motor domain and the ATPase activity. We hypothesize that conformational transitions in myosin gate ATP hydrolysis through regulating not only positions of specific amino acids in the active site but also the orientation and dynamics of water molecules surrounding the hydrolysis site. To verify and consolidate such a hypothesis, state-of-the-art molecular simulations are proposed to analyze the mechanism of ATP hydrolysis in different conformational states of the myosin II motor domain and relevant mutants; the simulations include classical molecular dynamics and combined QM/MM methods. The specific aims are: (1) Determine the catalytic mechanism of ATP hydrolysis in the closed state of the motor domain. (2) Determine if ATP hydrolysis is prohibited in the open state of the motor domain, and if so, identify key differences between the open and closed conformations that dictate the hydrolysis energetics. (3) Explain, in energetical and mechanistic terms, the roles of active site residues, which have been shown by mutagenesis studies to have various effects on ATP hydrolysis and motility. Myosin-ll was chosen because it is the only motor system that has high-resolution structures for multiple conformational states, and computational results can be compared with a large body of biochemical and biophysical data. The proposed simulation study will provide a framework for bridging experimental data from different disciplines to establish sensible theoretical models for mechanochemical coupling in myosin and other molecular motors; the microscopic insights will have a profound impact on our ability to design strategies to treating serious diseases caused by myosin dysfunction such as cardiomyopathy. The simulation work will be closely coupled to experimental studies through collaborations; the combination of structural, kinetic and motility data will provide the experimental tests necessary to verify and refine simulation techniques, which is of tremendous value to the field of computational enzymology.